Manuel Nuno Melo

Feature (un)wrap enhancement

16

Changes made in this Pull Request: - Optimizes unwrapping by pre-screening AG fragments that span more than half the box vectors; - Extends the universe-validated cache mechanism to avoid re-calculating...

Implements compact wrapping and nojump, and updates associated transforms.

11

Implemented function `distances.apply_compact_PBC`. With it I could easily adapt `AtomGroup.wrap()` to also do compact wrapping, depending on a keyword. Backward compatibility is maintained. Also optimized the way `compound_indices` are handled...

List of protein-defining resnames now read from force field.

6

The list of protein resnames is now parsed from the `protein_resnames` macro in the `.ff` files. This allows flexibility in what is processed as a protein. For instance, I created...

ChainReaders now apply their own transformations

13

Closes #3343 Takes into account over-transformation of `SingleFrameReaders`, which never re-create their `ts` when re-accessed. PR Checklist ------------ - [x] Tests? - [x] Docs? - [x] CHANGELOG updated? - [x]...

Unwrap with default settings fails for virtual sites

26

## Expected behavior ## For an `ag` containing virtual-sites (which are always massless in GROMACS topologies) `ag.unwrap()` should unwrap `ag` atoms. ## Actual behavior ## ``` Traceback (most recent call...

Parallel implementation not leveraging the lightweight serializations/de-serializations of MDA

6

Right now parallel analyses do their own sort of pickling/unpickling. While this gives the analysis more control and possibly skips unneeded pickle/unpickle cruft, it has a number of drawbacks: -...

When adding new lipids with the `-al*` flags, the tail names would get mangled: beads `C1A`, `C2A`, etc. become `CA1`, `CA2`, etc. It is a simple matter of swapped letter/number...